Mitochondria play an essential metabolic role in all eukaryotic cells because they are the organelles that provide energy to drive chemical reactions. Mitochondria are highly dynamic and change their shape between discrete structures and large interconnected networks by selective fission and fusion of their membranes. The resulting fused three-dimensional networks can permeate through an entire cell, but their advantages, relative to isolated discrete organelles, are not fully understood.  It has been suggested that networks might have the following functions: dampen biochemical fluctuations; provide a mechanism for quality control and selective mitophagy; enable diffusion of proteins; and transmit calcium signals and mitochondrial membrane depolarization throughout the cell [1, 2]. Despite their highly dynamic state, useful structural information can nevertheless be obtained from electron microscopy by studying the organization of mitochondrial networks at a given time point.

We have used a SIGMA VP (Zeiss Inc.) scanning electron microscope (SEM), equipped with a 3View (Gatan Inc.) serial block face (SBF) system [3], to quantify the connectivity of mitochondrial networks in entire insulin-secreting b-cells of mouse pancreatic islets of Langerhans. Previous work has suggested that alterations in mitochondrial fission and fusion might play a role in nutrient-induced b-cell apoptosis with possible involvement in the pathophysiology of Type 2 diabetes [4].

Previously, we found that manual segmentation of SBF-SEM mitochondrial volumes in a single b-cell took about one week for a trained operator, which precluded analysis of multiple cells within a reasonable time.  We have explored a faster and more practicable approach by making use of tools within the Amira visualization software package (FEI Inc.) to partially automate the segmentation process, and enabling an entire cell to be analyzed in approximately two hours (Figure 1). Using this approach we were able to analyze quantitatively mitochondrial networks in individual pancreatic b-cells in terms of the total mitochondrial volume, average volume per network, total network length, and average network length. We found that approximately one-third of the mitochondrial volume was contained in networks of length less than 3 µm, one-third in networks of length between 3 µm and 10 µm, and one-third in highly fused networks of length between 10 µm and 60 µm (Figure 2).

SBF-SEM thus provides quantitative data on the organization of mitochondrial networks, making it feasible to test computational models for mitochondrial fusion and fission.

The authors thank Drs. A.L. Notkins, T. Cai, and H. Xu for providing the specimens of pancreatic islets of Langerhans. This research was supported by the intramural program of the National Institute of Biomedical Imaging and Bioengineering, NIH.

[1]  H. Hoitzing et al, Bioessays 37 (2015) p. 687.

[2]  B. Clancy et al, Nature 523 (2015) p. 617.

[3]  W. Denk and H. Horstmann, PLoS Biol. 2 (2004) p. 1900.

[4]  A.J.A. Molina et al, Diabetes 58 (2009) p. 2303.

[5]  C.R. Pfeifer et al, J. Struct. Biol. 189 (2015) p. 44.

[6]  A. Shomorony et al, J. Microsc. 259 (2015) p. 155.

Figures:

Figure 1. Semi-manual segmentation of mitochondrial networks in four mouse pancreatic β-cells; each network is indicated in a separate color.

Figure 2. Distribution of mitochondrial networks as a function of network length for seven pancreatic β-cells; vertical red lines indicate the approximate network lengths containing one-third and two-thirds of the total mitochondrial volume.

« Back to The 16th European Microscopy Congress 2016

EMC Abstracts - https://emc-proceedings.com/abstract/study-of-mitochondrial-networks-in-pancreatic-islet-%ce%b2-cells-by-sbf-sem/